• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 63 subscribers
  • Views 680 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM3478 - some help and a few questions

Mehdi Chafik
Prodigy 60 points
Other Parts Discussed in Thread: LM3478

Hello everyone,

I'm working on a project  that will require me to step-up the voltage coming from a battery (3V to 5.5V) to 50Volts@300mA.

I'm using an LM3478 for that (sizing done by webench 3201.SEPIC.pdf)

A few remarks:

- The prototype i've done is powered by a power supply (up to 30V@3A), not a battery.

- I've read about the Vin problem (current limited power supply or power supply too slow to get to 2.97V). I've implanted Rbp and Cbp to counter that (like indicated in webench). I checked on the scope, power supply rising time is close to 200 us, so everything should be fine.

-Everything is similar to the webench sizing except that i couldn't find the MOS recommended. I've used an IRF1010 instead. I'm going to change it tomorrow with an IPB08CN10N. Rds, Vgs th and reverse capacitance are very good. Only the total gate charge is higher than the MOS recommended.

- I made this quick and dirty so no PCB but everything soldered on a prototype board. I minimized the AC loop. For the rest i did as i could.

Here is my problem:

- It's working right now if the input is between 16V and up to 24V (haven't tested higher voltage). The outpout is quite steady and always around 50 Volts.

- If the input voltage is between 13 and 15 Volts, the system is unstable. I first have a correct ouput value, it then stops (after a few seconds).

-If the input voltage is below 13 Volts, nothing works..i'm not in short circuit through the MOS..i just have Vout= 0 Volts. Vin on the chip is 13 Volts too.

-I've tried using a battery (approx 4V), result is the same. Nothing at the ouput, no short circuit due to the MOS.

Any idea why or how i could understand where the problem come from?

Thanks!

  • 0
    TI__Guru 72425 points

    Send us some waveforms when the system is working and just before it stops; switching voltage and voltage across current sense R.

    Does it stop switcing completly??  Is something getting too hot??

    I will pass this along to our boost experts.

    Frank D

     

  • 0 in reply to Frank De Stasi
    Prodigy 60 points

    Hi Frank. Thanks for your quick answer!

    First of all, i made a mistake in my last report. When i'm between 16V and 24V, everything works fine. I then lower the input voltage from the power supply. When i reach approx. 14V, the system stops but contrary to what i said, the DR output is set to high and i have then a shortcut. So it stops switching completely.

    I guess i know why as when i looked at the Vin pin on the scope, i saw that i had a huge dip close to 5V when the MOS is switching (orange line is the Vin and the blue line is the DR output). This voltage dip seems to be constant whatever the Vin(16 to 24V) or the load. I tried with a 1.6k resistance (300mA @50V) or without any load.

    So i guess that the Vin gets close to 3V for a brief moment and the circuit stops.

    I tried filtering that by changing Cin and Cbp without success.

    My power supply is able to deliver 3A@30V so what could be the problem? I tried changing the MOS. I used one with a better Rds and better Vgs th but it's worse!

    I also tried again with a battery (4Volts) and with a MOS driveable with a Vgs at 3V and i do have the DR output  high (no switching) and the battery voltage is obviously slowly dipping as it is in shortcut.  When i switch the product on, it takes less than 20us for the Vin voltage to settle at 4V.

    Not sure on what to do next. Any advice on what i should  try tweaking?

    BTW, nothing gets really hot. The MOS is a little bit hot (but nothing strange) when i outputting 300mA @50V

    Here are the waveforms you ask me.

    With Vin=16V.

    DR output

    R sense :

    With Vin=14V (the circuit stops a few seconds later):

    DR output:

    R sense:

  • 0 in reply to Mehdi Chafik
    TI__Guru 72425 points

    Yes, if the input dips to low the part will shut off.  The fact that the battery works means that

    you other input supply is somehow drooping.  I would try shorter and fatter input leads and are

    large electrolytic cap on the input.

    Frank D

     

  • 0 in reply to Frank De Stasi
    Prodigy 60 points

    Hello Frank,

    With the battery, the system DOESNT work! When i switch the product on, the voltage quickly settle to 4V but the circuit doesn't start correctly. The DR output stays high with no switching at all.

    I will try to shorten the input leads and tweaking again the input capacitor but would like your opinion on why the switching doesn't start when connected to a battery @4V.

    Thanks again for your help.

    Mehdi